Help center
Targeting & triggers
A/B testing your popups: a starter guide
When to test, what to test, and how to read the results without a statistics degree.
A/B testing is the single highest-leverage thing you can do once your popup is shipped. Here's a starter framework.
When to start testing
Wait until you have at least 500 impressions on your control popup. With less, you're testing noise.
What to test first
- Headline. Single biggest CVR driver. Try short vs descriptive, urgency vs benefit-led.
- Offer. 10% off vs free shipping vs gift with purchase. Often more impactful than copy.
- Trigger. Timed (5s) vs scroll (50%) vs exit-intent.
- Form length. Email-only vs email + name. Fewer fields almost always wins.
What to test later
- Visual design — image vs no image, single-column vs multi-column.
- Multi-step vs single-step.
- Mobile-specific variants (different copy, smaller form, etc).
Reading the results
Don't pick winners after 50 impressions. The dashboard surfaces a confidence indicator once each variant has hit a meaningful sample size — usually around 1,000–2,000 impressions per variant for a clear winner.
If two variants look identical after enough traffic, they probably are. Move on and test something more ambitious.